“Grey Goo” may end the world.

In the Grey Goo scenario, tiny, self-replicating nanobots deconstruct and consume every organic and inorganic material on Earth, multiplying at an exponential rate until everything is just a homogenous, frolicking mass of tiny bots.1

The Grey Goo scenario can be classified as theoretically possible, but highly, highly improbable. In comparison, it’s more likely that the bacteria that live harmlessly inside your nose mutate and kill you. As a scientist, the last thing I want to be part of is fearmongering against a technology that can be very beneficial, and I’ll say right off the bat that if we are capable of doing something, we are most likely capable of controlling or stopping it, if not outright reversing it.

Still, it is always fun look at the “what if” scenarios, and it doesn’t hurt to draw the regulatory line between A-OK total madness.

So, back to Grey Goo. How do we get to this point?

Consider cells, the basic building block of every living thing on this planet (and possibly other planets too, let’s not rule that out). It’s odd to think about cells in a non-anthromorphogenic way when considering the cell cycle. Cells have their own metabolism; they consume, reproduce, die, and are turned over. Barring any sort of external cause like tissue injury, a cell is responsible for its own demise, as it commits suicide through a process called apoptosis. Apoptosis and its various mechanisms are still a topic of study, however major hallmarks include a mass release of enzymes that hack apart the cell from the inside-out.2

As it goes, cell death is essential for the organism’s survival. As cells grow and divide, the risk for small errors in DNA sequence increases, and a very common small error may result in the cell not murdering itself. If this normal growth-multiply-death process is disrupted, cell division may go unchecked, and the buildup of abnormal cells becomes a tumor.2 In the worst-case scenario, whatever cellular dysfunction is occurring results in rapid, uncontrolled growth, which we classify as malignant. In the even worstess-case scenario, bits and pieces of this malignant tumor break off and enter the bloodstream, which in turn expresses them to various other parts of the body where they continue to grow and divide. This is called metastatic cancer, and it will kill you dead.

There is no illness quite like cancer. There is no specific pathogen that can be eliminated by our immune system supplemented by medical intervention. Our own cells that have run afoul, many different things having gone wrong to result in unchecked cell growth. For years, cancer treatments have relied on pumping cytotoxic drugs that inhibit cell growth through the body (chemotherapy). As expected, chemotherapies have terrible side effects that may last well beyond the point of remission, caused by off-target damage to normal cells.

This is where new drug delivery systems using nanoparticles come in. The idea is to specifically target the abnormal cells, and directly deliver a lethal dose of the cancer drug.

For targeted drug delivery, we need a minimum of three parts: First, an antibody to find the cancer cells, second, a drug to kill the cancer cells, and third, a “vehicle” that delivers the antibody and the drug to their destinations. The details amaze; antibodies are raised against the specific cancer cells and act as the “driver” to tag and attach to cancer cells, and is bound to a nanoparticle of gold, iron, or a biological-based molecule which acts as the delivery vehicle for the drug. The end result is that we bypass any innocent bystanders along the way. These methods have already been put into action in many different cases, including brain cancers, basal cell carcinomas, leukemia, and cancers of the bone, among others.3 Delivery of cancer drugs is not the only application, nanoparticles have been shown to be able to dissolve blood clots, repair damaged brain tissue, and clear out bacterial toxins.45

New technologies have allowed us to work at the nanoscale, opening up a ton of opportunities to solve Big Problems, like cancer. Other Big Problems tackled by nanotechnology involve uses in renewable energy, and construction of new building materials. A handy emerging use of nanotechnology is through remediation, or the cleanup and treatment of toxic leavings from groundwater and soils, including from oil spills, petroleum, heavy metals such as cadmium and arsenic, and possibly even radioactive waste.6 Remediation has already been practiced for some number of years by bacteria, fungi, and even plants (phytoremediation), however nanoparticles are also great at it due to their large surface area per unit mass, which makes them both really, really effective, and tiny enough to move through sediment.

It’s one thing to be able to make something possible in the sciences. It’s entirely different to be able to afford it. Can we afford the cost of the materials and personnel to develop the highly specific targeted drug delivery systems, not mentioning the cost of the research and development of these products? Upon the development of new technologies, there is always a rush to find ways to scale up, bring down the cost, and speed up the process so that implementation of these technologies into our daily lives is economically feasible.

There are two major classifications of remediation, In situ, and Ex situ. Remediation was traditionally done ex situ, where contaminated groundwater was pumped out of aquifers for treatment, then put back into the system all nice and clean. Nanoparticles can actually get in between sediment, making it unnecessary to pump the contaminated water out prior to treatment. As expected, skipping the pump and transportation step cuts down on the cost of remediation, and this method is absolutely feasible for treatment of pollution. As of 2014, there are dozens of In situ nanoremediation projects in place around the world.7

Now, let’s think about a hypothetical situation in which we are going to need a lot of nanoparticles, quickly. Right now many of these particles are constructed via assembly line, which is efficient, but has its limits. Let’s say that there was a massive explosion on an off shore oil drill, resulting in a continuous gushing of oil into a formerly pristine and economically valuable body of water. Another situation (still absolutely, totally, hypothetical, of course) is a large chemical spill into a major river, which will quickly result in a toxic widespread distribution of this particular chemical into populated areas. Constantly building nanoparticles and releasing them may not stop or slow the spread of pollution in time before the damage hits; but possibly, if we can use nanobots capable of self-replication then perhaps that will give us large enough numbers to control the toxic mess.

This is where the Grey Goo scenario enters the realm of the theoretically possible.

When we think about self-replication, we can draw inspiration and a possible design based on biological systems. Cells are capable of self-replication through ribosomes, which are cellular assemblies that link the basic building blocks of proteins (amino acids) together based on the order specified by the genetic instructions (messenger RNA). Artificial self-replication is a little tricky; in biological systems there are numerous complex mechanisms to deliver the specific and necessary components at the right time, to the right place within a cell. To self-replicate, a machine would need to be able to move specific chemical reactions by positioning reactive molecules at the right time, with atomic precision.8 Despite these difficulties, research groups have had some success with developing machinery with some capability into self-assembly or self-replication.

Let’s say that we wanted to go even further, and develop nanomachines that could not only self-replicate, but also adapt to a changing environment, with a goal of being able to remediate a number of different types of pollutants and/or replicate under different environmental conditions with variety of materials available. The major driving force for evolution in biological systems is natural selection. If we had self-replicating nanobots able to adapt to their environment, they would have the capability to undergo natural selection, and evolve independently of their fleshy creators.

At this point, natural selection in artificial life (artificial natural selection??) is still very theoretical, and starts nudging this column into the realm of science fiction.

Sorry about that.

Let’s look at our end product. We have a batch of nanobots capable of taking up and consuming a wide variety of molecules, which self-replicate at an exponential rate. I’d say that at this point, we definitely have gone too far, and are headed towards that Grey Goo scenario. It’s doubtful that we will reach this point, because of regulatory protocols and risk assessment analyses that will keep this sort of nanobot from being released into the environment.

But, if we do end up with a rogue nanobot that increases exponentially and consumes everything in its path, then I suppose our options are to kill it with fire, or nuke it from orbit. Or, if we don’t want to cause too much damage to the surviving, surrounding environment, we can find a way to target and deactivate just the rogue nanobots.

This all sounds pretty familiar.

- - -

Thanks to computer scientist extraordinaire Dr. Shawn O’Neil for the helpful comments!

- - -

1 Drexler, K.E. 1986. Engines of Creation: The Coming Era of Nanotechnology. Anchor Books, NY, USA

2 Wong, R.S. 2011. Apoptosis in cancer: Pathogenesis to treatment. J Exp Clin Cancer Res. doi: 10.1186/1756-9966-30-87

3 Targeted Therapies Fact Sheet:http://www.cancer.gov/about-cancer/treatment/types/targeted-therapies/targeted-therapies-fact-sheet

4 Voros, E et al., 2015. TPA Immobilization on Iron Oxide Nanocubes and Localized Magnetic Hyperthermia Accelerate Blood Clot Lysis. Advanced Functional Materials. DOI: 10.1002/adfm.201404354

5 Kaittanisa, C., Santraa, S., Perez, JM. 2011. Emerging nanotechnology-based strategies for the identification of microbial pathogenesis. Advanced Drug Delivery Reviews 62:408-423

6 Mallampati, SR. et al., 2013. Novel Approach for the Remediation of Radioactive Cesium Contaminated Soil with nano-Fe/Ca/CaO Dispersion Mixture in Dry condition. E3S Web of Conferences 1, 08003. DOI: 10.1051/e3sconf/20130108003

7 U.S. Environmental Protection Agency

8 Phoenix, C., Drexler, KE. 2004. Safe exponential manufacturing. Nanotechnology 15:869-872